Page 216 - 2024-Vol20-Issue2
P. 216
212 | Gochhait, Sharma & Bachute
weekday, and time. These factors have been identified to exert power demand. On the other hand, relative humidity at 2 me-
a substantial influence on electricity consumption patternsand ters(RH2M) and wind speed at 10 meters (WS10M) demon-
are leveraged to enhance the precision of load forecasting strate weaker correlations with power demand. The correla-
models. By incorporating these time indicators into the mod- tion analysis demonstrates that weather parameters have vary-
eling process, more accurate predictions of electricity demand ing degrees of influence on power demand in both Odisha and
can be achieved. Load Parameters: In this research, particular Delhi. Notably, relative humidity consistently shows a signifi-
emphasis is placed on load parameters, particularly the previ- cant negative correlation with power demand, underscoring its
ous half-hourly load in MW. The analysis of historical load importance in load forecasting. Meanwhile, temperature, wet
data enables the identification of patterns and trends that are bulb temperature, and specific humidity exhibit strong positive
instrumental in developing accurate load forecasting models. correlations, highlighting their substantial impact on power
By considering this load parameter, valuable insights can be demand. It is crucial to note that even weather parameters with
gained to enhance the precision of load forecasting. weaker correlations, such as dew/frost point and wind speed,
The ultimate objective of this research study, focusing on should not be disregarded, as they can still contributeto power
a comparative long- term electricity forecasting analysis of demand variations. In conclusion, this correlation analysis
Odisha and Delhi states, is to enhance our comprehension emphasizes the significance of considering weather parame-
of electricity load consumption patterns. By delving into ters when examining power demand. The unique relationships
the intricacies of these patterns, the research endeavors to revealed by the correlation coefficients offer valuable insights
contribute to the advancement of load forecasting models, ulti- for developing energy management strategies and more accu-
mately leading to more efficient and sustainable management rate load forecasting models. Furthermore, Table IV provides
of power systems. The insights gained from this analysis will a direct comparison of the correlation coefficients between
serve as a valuable resource for decision-makers in the energy weather parameters and power demand in Odisha and Delhi.
sector, enabling them to make informed choices anddevise The results indicate that in Odisha, there are positive correla-
strategies that optimize resource allocation and ensure a reli- tions for T2M, T2MDEW, T2MWET, QV2M, and WS10M,
able power supply. although with relatively weak magnitudes. On the other hand,
b: Correlation of data RH2M exhibits a negative correlation. In Delhi, positive cor-
The comparison of analysis of correlation Understanding the relations are observedfor T2M, T2MDEW, T2MWET, QV2M,
interplay between weather parameters and power demand is and RH2M, with T2MWET showing the strongest correlation
vital for comprehending electricity consumption patterns. In coefficient. However, WS10M demonstrates a weak positive
this study, we conducted a correlation analysis to investigate correlation.
the relationship between selected weather param- eters and Comparing the correlations between Odisha and Delhi, it
power demand in the states of Odisha and Delhi. The corre- becomes evident that Delhi generally exhibits stronger cor-
lation coefficientsprovide valuable insights into the strength relations for most weather parameters, emphasizing a more
and direction of these relationships. The interpretation of significant influence of these parameters on power demand in
correlation coefficients, presented in Table II for the state of Delhi compared to Odisha.
Odisha, reveals that temperature at 2 meters (T2M), wet bulb Overall, this comparison underscores the importance of adopt-
temperature at 2 meters. ing region-specific approaches to better comprehend the rela-
(T2MWET), and specific humidity at 2 meters (QV2M) ex- tionship between weather parameters and power demand in
hibit positive correlations with power demand. This implies different states.
that an increase in these parameters corresponds to higher c: Preprocessing of Data
power demand. Conversely, relative humidity at 2 meters
(RH2M) shows a negative correlation, suggesting that as rela- TABLE II.
tive humidity increases, power demand tends to decrease. The INTERPRETATION OF CORRELATION FOR THE
correlations for other parameters, such as dew/frost point at 2
meters (T2MDEW) and wind speed at 10 meters (WS10M), PARAMETERS OF ODISHA STATE
are relatively weaker.
Similarly, Table III provides the interpretation of correlation Parameter Correlation Coefficient
coefficients for the stateof Delhi. In Delhi, wet bulb tem- T2M 0.153685807
perature at 2 meters (T2MWET), the temperatureat 2 meters 0.08891532
(T2M), and specific humidity at 2 meters (QV2M) exhibit T2MDEW 0.130545477
strong positive correlations with power demand, indicating T2MWET -0.0614450
that an increase in these parameters is associated with higher 0.094774887
RH2M 0.098106045
QV2M
WS10M
